CN111461197A - Spatial load distribution rule research method based on feature extraction - Google Patents

Spatial load distribution rule research method based on feature extraction Download PDF

Info

Publication number
CN111461197A
CN111461197A CN202010231648.5A CN202010231648A CN111461197A CN 111461197 A CN111461197 A CN 111461197A CN 202010231648 A CN202010231648 A CN 202010231648A CN 111461197 A CN111461197 A CN 111461197A
Authority
CN
China
Prior art keywords
load
land
density
electricity utilization
clustering
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202010231648.5A
Other languages
Chinese (zh)
Inventor
彭鹏
邵宇鹰
唐轶斐
郑申辉
姚初晴
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
State Grid Shanghai Electric Power Co Ltd
Original Assignee
State Grid Shanghai Electric Power Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by State Grid Shanghai Electric Power Co Ltd filed Critical State Grid Shanghai Electric Power Co Ltd
Priority to CN202010231648.5A priority Critical patent/CN111461197A/en
Publication of CN111461197A publication Critical patent/CN111461197A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
    • G06F18/24133—Distances to prototypes
    • G06F18/24137—Distances to cluster centroïds
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply

Landscapes

  • Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Economics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Probability & Statistics with Applications (AREA)
  • Water Supply & Treatment (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Marketing (AREA)
  • Primary Health Care (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention provides a method for researching a space load distribution rule based on feature extraction, which comprises the following steps: s1, generating a plurality of electricity utilization functional cells in a certain area according to actual land conditions and researching the power load to obtain a load sample; s2, extracting geographic feature information of each electricity utilization function cell, giving a land type, and calculating the load density of each electricity utilization function cell according to the land type and the power load data; s3, clustering all electricity utilization function cells into N types by taking the geographic characteristic information as input information; s4, extracting the load density distribution characteristics of the electricity utilization functional cells of various types and various land types; and S5, obtaining the space load distribution rule of the region according to the load density distribution characteristics. The advantages are that: the method considers the geographical characteristic difference of the power load and the characteristics of partitioning and typing of the power load, effectively delineates the distribution of the space load, can provide reliable basis for planning and scheduling of the power distribution network, and has strong practicability.

Description

Spatial load distribution rule research method based on feature extraction
Technical Field
The invention relates to the field of research on spatial load distribution rules, in particular to a method for researching the spatial load distribution rules based on feature extraction.
Background
With the rapid advance of urbanization and industrialization processes in China, the electricity consumption of the society increases year by year, the scale of the power distribution network is continuously enlarged, and new requirements are provided for the management and planning of the power distribution network. In order to realize the fine management of the power distribution network and the optimal allocation of resources and improve the quality and speed of the power distribution network planning, the distribution rule of the space load needs to be deeply researched. At present, the research on the spatial load distribution rule is less, and the existing research on the spatial load distribution rule does not fully consider the geographical feature difference and the discontinuous characteristics of the spatial load distribution.
The spatial load distribution rule is different from the spatial distribution rule of the categories of atmospheric pollution, geological disasters, natural resources and the like, and the distribution of the spatial load is not only related to geographic factors, but also related to factors such as land types, population density and the like. The distribution of the space load generates a large difference due to the land type, the land area and the land area, and the distribution in the space is discontinuous, so that the factors such as the land type and the regional difference need to be considered when the space load distribution rule is obtained, and a traditional interpolation analysis method cannot be used, however, researchers do not pay much attention to the method at present.
Disclosure of Invention
The invention aims to provide a method for researching the space load distribution rule based on feature extraction. The method considers the characteristics of geographical feature difference and blocking and type division of the power load, effectively delineates the distribution of the space load, and has strong practicability.
In order to achieve the purpose, the invention is realized by the following technical scheme:
a space load distribution rule research method based on feature extraction comprises the following steps:
s1, generating a plurality of electricity utilization function cells in a certain area according to actual situations, and researching the power load of each electricity utilization function cell to obtain a plurality of load samples;
s2, extracting geographic feature information of each electricity utilization function cell, giving the land type of each electricity utilization function cell, and calculating the load density of each electricity utilization function cell according to the land type information and the power load data;
s3, clustering the electricity utilization function cells into N types by taking the geographic characteristic information in the step S2 as input information;
s4, extracting the load density distribution characteristics of the electricity utilization functional cells of various types and various land types;
and S5, obtaining the space load distribution rule of the whole region according to the load density distribution characteristics extracted in the step S4.
Preferably, the geographic characteristic information comprises directly acquired information and information to be calculated,
the directly acquired information includes: distance a1 from nearest main road, distance a2 from nearest business center, distance A3 from nearest residential district, distance a4 from nearest industrial district, distance a5 from nearest school, distance a6 from city center or district center, distance a7 from nearest river, distance A8 from nearest train station;
the information that needs to be calculated includes: residential user density factor a9 in a radius of the surrounding 2 cells, commercial user density factor a10 in a radius of the surrounding 4 cells, and industrial user density factor a11 in a radius of the surrounding 6 cells.
Preferably, the user density factor calculation formula is:
Figure BDA0002429454970000021
in the formula, G is the density factor of a certain type of user, U is the number of the certain type of user, S is the area, and j is the type.
Preferably, in the step S2,
the land types of the electricity utilization function cells comprise commercial land B, industrial land M, residential land R and public management and public service land A,
wherein, industrial area M contains: production workshops, storehouses and auxiliary facility land areas of industrial and mining enterprises;
the residential land R includes: the land of the residence and the corresponding service facilities;
the land for public management and public service A comprises: the land for administrative, cultural, educational, sports, health institutions and facilities, excluding the land for service facilities in the residential land;
commercial site B contains: the land for commercial, commercial and entertainment sports facilities does not include the land for service facilities in residential land and the land for public administration and public service units.
Preferably, in step S3:
and clustering the electricity utilization functional cells into N categories by adopting a fuzzy C-means clustering algorithm.
Preferably, the step S3 is specifically:
s31, initializing various parameters, which specifically include: converting the geographical features of the electricity utilization functional cells into a data set X, dividing the data set X into c types by fuzzy clustering, including c clustering centers, setting the iteration number h to be 0, and setting an iteration stop condition;
s32, sequentially calculating Euclidean distances from all load samples in the data set X to all clustering centers, and calculating sample membership degrees of all samples and all clustering centers;
s33, updating the clustering center;
s34, calculating an objective function;
s35, determining whether an iteration stop condition is satisfied, if yes, going to step S36, if no, making h equal to h +1, and going to step S32;
and S36, outputting the clustering center and the clustering result.
Preferably, the step S31 is specifically:
and converting the geographic characteristics of the electricity utilization functional cells into a data set X { X ═ X by taking the geographic characteristic information as input information1,x2,L,xnN is the number of load samples, i.e. the number of electrically functional cells, each load sample contains a elements, the kth load sample xk＝{xk1,xk2,L,xkaFuzzy clustering divides a data set into c types, wherein the set of c clustering centers is as follows: z ═ Z1,z2,L,zcThe ith cluster center is zi＝{zi1,zi2,L,ziaSetting the iteration time h as 0, and setting an iteration stop condition;
in step S32, the euclidean distance between the kth load sample and the ith cluster center
Figure BDA0002429454970000031
Comprises the following steps:
Figure BDA0002429454970000032
sample membership of kth load sample and ith cluster center
Figure BDA0002429454970000033
Is composed of
Figure BDA0002429454970000034
Wherein the content of the first and second substances,
Figure BDA0002429454970000035
the euclidean distance of the kth load sample to the jth cluster center,
Figure BDA0002429454970000036
Figure BDA0002429454970000037
in step S33, each cluster center is iterated according to formula (4),
Figure BDA0002429454970000041
in step S34, the objective function f (h) is:
Figure BDA0002429454970000042
preferably, in the step S4,
and extracting load density distribution characteristics by using a nonparametric kernel density estimation method.
Preferably, the step S4 specifically includes:
s41, estimating nonparametric nuclear density;
and S42, extracting load density distribution characteristics according to the non-parameter kernel density estimation.
Preferably, the kernel of the non-parametric kernel density estimation probability density function f (m) in step S41 is estimated as
Figure BDA0002429454970000043
Wherein n is the number of load samples, i.e. the number of electrically functional cells, m1,m2,L,mnFor the load density value, m, of each load sampleiThe load density value of the ith load sample.
Compared with the prior art, the invention has the following advantages:
(1) according to the method for researching the spatial load distribution rule based on the feature extraction, a plurality of electricity utilization functional cells are generated according to the actual land of a certain area, the geographic feature information of the electricity utilization functional cells is extracted, then the electricity utilization functional cells are classified by applying a clustering algorithm, typical distribution features of sample load densities of different types and different land types are extracted by using a non-parameter kernel density estimation method, and further the distribution rule of each type of spatial load in the research area is obtained;
(2) according to the characteristic extraction-based space load distribution rule research method, clustering analysis is carried out according to different geographic characteristics and population density factors, power utilization function cell clusters with different types are formed, the land types of the power utilization function cell clusters are given according to standards, and an effective power utilization function cell load classification method is provided for a power distribution network.
Drawings
FIG. 1 is a schematic diagram of a method for studying a spatial load distribution rule based on feature extraction according to the present invention;
FIG. 2 is a schematic diagram of the fuzzy C-means clustering algorithm of the present invention;
FIG. 3 is a probability density curve of a load density of a typical load sample of the present invention;
FIG. 4 is a selected profile of a study area in accordance with an embodiment of the present invention;
fig. 5 is a histogram of spatial load distribution rules according to an embodiment of the present invention.
Detailed Description
The present invention will now be further described by way of the following detailed description of a preferred embodiment thereof, taken in conjunction with the accompanying drawings.
As shown in fig. 1, a method for studying spatial load distribution rules based on feature extraction according to the present invention includes the following steps:
and S1, generating a plurality of electricity utilization function cells in a certain area according to actual situations, and widely investigating the power load of each electricity utilization function cell to obtain a plurality of load samples with sufficient quantity and comprehensiveness.
And S2, extracting the geographic characteristic information of each electricity utilization function cell, giving the land type of each electricity utilization function cell, and calculating the load density of each electricity utilization function cell according to the land type information and the power load data.
The geographic feature information includes directly acquired information and information to be calculated, and the directly acquired information includes: distance a1 from nearest main road, distance a2 from nearest business center, distance A3 from nearest residential district, distance a4 from nearest industrial district, distance a5 from nearest school, distance a6 from city center or district center, distance a7 from nearest river, and distance A8 from nearest train station.
The information to be calculated comprises: the residential user density factor a9 in the radius range of the surrounding 2 cells, the commercial user density factor a10 in the radius range of the surrounding 4 cells, and the industrial user density factor a11 in the radius range of the surrounding 6 cells, in the present embodiment, the user density factor is calculated as:
Figure BDA0002429454970000051
in the formula, G is the density factor of a certain type of user, U is the number of the certain type of user, S is the area, and j is the type.
In addition, in the step S2, the land types of the electricity utility cells include a commercial land B, an industrial land M, a residential land R, and a public management and public service land a. Wherein, industrial area M contains: production workshops, storehouses and accessory facilities of industrial and mining enterprises; the residential land R includes: the land of the residence and the corresponding service facilities; the land for public management and public service A comprises: the land of the institutions and facilities such as administration, culture, education, sports, sanitation and the like, excluding the land of the service facilities in the residential land; commercial site B contains: the facilities land for various businesses, entertainment, health and the like does not include the service facility land in the residential land and the public management and public service land.
And S3, clustering the electricity utilization function cells into N types by using the geographic characteristic information in the step S2 as input information. In this embodiment, each electricity consuming functional cell is clustered into N categories by using a fuzzy C-means clustering algorithm.
As shown in fig. 2, the step S3 specifically includes:
s31, initializing various parameters, which specifically include: and converting the geographic characteristics of the electricity utilization functional cells into a data set X { X ═ X by taking the geographic characteristic information as input information1,x2,L,xnN is the number of load samples, i.e. the number of electrically functional cells, each load sample contains a elements, the kth load sample xk＝{xk1,xk2,L,xkaFuzzy clustering divides a data set into c types, wherein the set of c clustering centers is as follows: z ═ Z1,z2,L,zcThe ith cluster center is zi＝{zi1,zi2,L,zia}. For example, a ═ 11, xk＝{xk1,xk2,L,xkaAnd 11 geographic characteristic indexes of the electricity utilization functional cell are obtained. The clustering center number c and the iteration stop condition e are set, and an iteration counter, that is, the iteration number h is set to 0. Randomly selecting c samples in a data set X, and assigning the c sample values to an initial clustering center Z ═ Z(h)。
And S32, sequentially calculating Euclidean distances from all load samples in the data set X to all the clustering centers, and calculating the sample membership degrees of all the samples and all the clustering centers.
Wherein, the Euclidean distance from the kth load sample to the ith cluster center
Figure BDA0002429454970000061
Comprises the following steps:
Figure BDA0002429454970000062
sample membership of kth load sample and ith cluster center
Figure BDA0002429454970000063
Is composed of
Figure BDA0002429454970000064
Wherein the content of the first and second substances,
Figure BDA0002429454970000065
the euclidean distance of the kth load sample to the jth cluster center,
Figure BDA0002429454970000066
Figure BDA0002429454970000067
and S33, updating the clustering center. The method specifically comprises the following steps: each cluster center is iterated according to formula (4),
Figure BDA0002429454970000071
and S34, calculating an objective function. The objective function F (h) is:
Figure BDA0002429454970000072
s35, determining whether the iteration stop condition is satisfied, if yes, going to step S36, if no, making h equal to h +1, and going to step S32. In this embodiment, it is determined whether | F (h +1) -F (h) | < e is satisfied, and if so, the process proceeds to step S36, and if not, the process proceeds to step S32 with h ═ h + 1.
And S36, outputting the clustering center and the clustering result. In this embodiment, sample x is sampledkDegree of membership u attributed to sampleikAnd recording the classification result of each sample in the largest class.
And S4, extracting the load density distribution characteristics of the electricity utilization functional cells of various types and various land types. In this embodiment, the load density distribution characteristics are extracted by a non-parametric kernel density estimation method.
The step S4 specifically includes: and S41, estimating nonparametric nuclear density.
In the step S41, the kernel of the non-parametric kernel density estimation probability density function f (m) is estimated as
Figure BDA0002429454970000073
Wherein n is the number of load samples, i.e. the number of electrically functional cells, m1,m2,L,mnFor the load density value, m, of each load sampleiThe load density value of the ith load sample.
As shown in fig. 3, a typical probability density curve of load density of a load sample is shown. And drawing a probability density curve of all sample load densities of the functional cells of a certain type and a certain land type according to the nonparametric kernel density function.
And S42, extracting load density distribution characteristics according to the non-parameter kernel density estimation.
In the embodiment, according to the probability density function curve, firstly, the load density data with the probability density lower than 0.004 is removed, and the influence of extreme values and atypical load density data is reduced; then, segmenting according to the probability density value, and dividing the probability density value into a low-section density section and a high-section density section or a low-section density section, a middle-section density section and a high-section density section, and reflecting the numerical difference of load densities caused by different scales; then, the load density value with the highest probability density is taken from the density of each section as the typical value of the load density of the section, and the distribution of the load density of the section in the neighborhood of the value is most concentrated, so that the load density value is most representative.
By applying the method, the space load density distribution characteristics of each category and each place type are extracted.
And S5, obtaining the space load distribution rule of the whole area according to the load density distribution characteristics extracted in the step S4.
Specifically, a spatial load distribution rule histogram is drawn based on the spatial load density distribution characteristics extracted in step S4, and the spatial load distribution rule of the entire region is obtained.
In an embodiment of the invention, a certain area with mature development and diversified land is selected for extracting the spatial load distribution rule. The selected area is 85km2519 electricity functional cells are generated from the research area as shown in FIG. 4, wherein the number of the commercial area cells is 96 and the number of the residential area cells is one, according to step S1The number of the land used cells is 163, the number of the industrial land used cells is 123, and the number of the common land used cells is 137.
And extracting the relevant spatial information of each electricity utilization function cell according to the method of the step S2, calculating the required density factor, and giving the land types of each electricity utilization function cell. As shown in table 1, some samples are listed in the table, and a1 to a11 are the spatial geographic feature information of the electricity-using functional cell.
TABLE 1 space geographic characteristic information of partial electricity utilization function cell
Figure BDA0002429454970000081
According to the step S3, the values of the electricity consuming functional cells a1 to a11 are clustered by using a fuzzy C-means clustering algorithm, as shown in table 2, which is a clustering result in this embodiment.
TABLE 2 clustering results of electricity utility cells
Figure BDA0002429454970000082
Figure BDA0002429454970000091
According to the clustering result in the table 2, each electricity utilization functional cell can be divided into three major categories, wherein the first category is a functional cell which is far away from a city center, a business center and a school, is close to an industrial area and a river and has low resident density factor; the second type is a functional district which is far away from stations and rivers, close to city centers, business centers, main roads and schools and has high merchant density factors; the third category is the functional community far away from the industrial area and close to the residential area, which has high residential density factor.
And according to the step S4, carrying out non-parameter nuclear density estimation on the load densities of the electricity utilization functional cells of different types and different land types, drawing a corresponding curve and extracting the distribution characteristics in the load density probability curve. The results of characteristic distribution feature extraction for each type of load density are shown in table 3.
TABLE 3 typical distribution characteristic extraction results of type load density for each land
Figure BDA0002429454970000092
As shown in fig. 5, according to step S5, a spatial load distribution rule histogram is drawn according to the extracted load density distribution characteristics of the electricity consumption functional cells of different types and different land types. As can be seen from table 3 and fig. 5, for the commercial land loads in different categories, the feature values of the two types of commercial load densities are higher than those of the one and three types, and the difference between the feature values of the load densities of the one and three types is not large. For industrial land loads in different classes, the characteristic values of each of the first, second and third classes tend to decrease, and the load density characteristic value in one class is significantly higher than that in the other two classes. For the resident loads in different categories, all the characteristic values of the resident load density of one category are lower than those of the second category and the third category, and all the characteristic values of the resident load densities of the second category and the third category are approximately equivalent. For the common loads in different classes, the typical values of the medium and low load density sections of the first, second and third classes of common function cells are not greatly different, but the typical value of the high load density section of one class is obviously lower than those of the second and third classes.
As can be seen from the combination of Table 3 and FIG. 5, the load density of the commercial land and the industrial land is obviously higher than that of the residential land and the public land by comparing the load of the different types of land in the transverse direction; the medium and low load density sections of the public land are typically lower than the residential land, but the high load density section is typically higher than the residential land. The commercial land is intensive, the industrial scale is large, the productivity is high, and therefore, the load density of the commercial land and the industrial land is high. The land occupation area of the leisure entertainment, cultural historic site and the like forming the middle and low load density sections of the public land is large, and the electric equipment is few, so the land occupation area is lower than the typical value of the middle and low load density sections of the residential land; the land used for medical sanitation, administrative office, education and scientific research and the like forming the public land high-load density section has a plurality of instruments and equipment, is long in starting time and high in synchronous rate, and is higher than the typical value of the high-load density section of the residential land.
In summary, according to the characteristic extraction-based spatial load distribution rule research method, a plurality of electricity utilization functional cells are generated according to the actual land of a certain area, the geographic characteristic information of the electricity utilization functional cells is extracted, then the electricity utilization functional cells are classified by applying a clustering algorithm, typical distribution characteristics of sample load densities of different types and different land types are extracted by using a non-parameter kernel density estimation method, and further the distribution rule of each type of spatial load in the research area is obtained. The method considers the characteristics of geographical feature difference and blocking and type division of the power load, effectively delineates the distribution of the space load, and has strong practicability.
While the present invention has been described in detail with reference to the preferred embodiments, it should be understood that the above description should not be taken as limiting the invention. Various modifications and alterations to this invention will become apparent to those skilled in the art upon reading the foregoing description. Accordingly, the scope of the invention should be determined from the following claims.

Claims (10)

1. A method for researching space load distribution rule based on feature extraction is characterized by comprising the following steps:
s1, generating a plurality of electricity utilization function cells in a certain area according to actual situations, and researching the power load of each electricity utilization function cell to obtain a plurality of load samples;
s2, extracting geographic feature information of each electricity utilization function cell, giving the land type of each electricity utilization function cell, and calculating the load density of each electricity utilization function cell according to the land type information and the power load data;
s3, clustering the electricity utilization function cells into N types by taking the geographic characteristic information in the step S2 as input information;
s4, extracting the load density distribution characteristics of the electricity utilization functional cells of various types and various land types;
and S5, obtaining the space load distribution rule of the whole region according to the load density distribution characteristics extracted in the step S4.
2. The method as claimed in claim 1, wherein the geographic feature information includes directly obtained information and information to be calculated,
the directly acquired information includes: distance a1 from nearest main road, distance a2 from nearest business center, distance A3 from nearest residential district, distance a4 from nearest industrial district, distance a5 from nearest school, distance a6 from city center or district center, distance a7 from nearest river, distance A8 from nearest train station;
the information that needs to be calculated includes: residential user density factor a9 in a radius of the surrounding 2 cells, commercial user density factor a10 in a radius of the surrounding 4 cells, and industrial user density factor a11 in a radius of the surrounding 6 cells.
3. The method for studying spatial load distribution rules based on feature extraction as claimed in claim 2, wherein the user density factor calculation formula is:
Figure FDA0002429454960000011
in the formula, G is the density factor of a certain type of user, U is the number of the certain type of user, S is the area, and j is the type.
4. The method for studying spatial load distribution rules based on feature extraction as claimed in claim 1, wherein in said step S2,
the land types of the electricity utilization function cells comprise commercial land B, industrial land M, residential land R and public management and public service land A,
wherein, industrial area M contains: production workshops, storehouses and auxiliary facility land areas of industrial and mining enterprises;
the residential land R includes: the land of the residence and the corresponding service facilities;
the land for public management and public service A comprises: the land for administrative, cultural, educational, sports, health institutions and facilities, excluding the land for service facilities in the residential land;
commercial site B contains: the land for commercial, commercial and entertainment sports facilities does not include the land for service facilities in residential land and the land for public administration and public service units.
5. The method for studying spatial load distribution rules based on feature extraction as claimed in claim 1, wherein in said step S3:
and clustering the electricity utilization functional cells into N categories by adopting a fuzzy C-means clustering algorithm.
6. The method for studying spatial load distribution rules based on feature extraction as claimed in claim 5, wherein said step S3 specifically comprises:
s31, initializing various parameters, which specifically include: converting the geographical features of the electricity utilization functional cells into a data set X, dividing the data set X into c types by fuzzy clustering, including c clustering centers, setting the iteration number h to be 0, and setting an iteration stop condition;
s32, sequentially calculating Euclidean distances from all load samples in the data set X to all clustering centers, and calculating sample membership degrees of all samples and all clustering centers;
s33, updating the clustering center;
s34, calculating an objective function;
s35, determining whether an iteration stop condition is satisfied, if yes, going to step S36, if no, making h equal to h +1, and going to step S32;
and S36, outputting the clustering center and the clustering result.
7. The method for studying spatial load distribution rules based on feature extraction as claimed in claim 6, wherein said step S31 specifically comprises:
and converting the geographic characteristics of the electricity utilization functional cells into a data set X { X ═ X by taking the geographic characteristic information as input information1,x2,L,xnN is a load sampleThe number is the number of the instant electric functional cell, each load sample comprises a elements, and the kth load sample xk＝{xk1,xk2,L,xkaFuzzy clustering divides a data set into c types, wherein the set of c clustering centers is as follows: z ═ Z1,z2,L,zcThe ith cluster center is zi＝{zi1,zi2,L,ziaSetting the iteration time h as 0, and setting an iteration stop condition;
in step S32, the euclidean distance between the kth load sample and the ith cluster center
Figure FDA0002429454960000031
Comprises the following steps:
Figure FDA0002429454960000032
sample membership of kth load sample and ith cluster center
Figure FDA0002429454960000033
Is composed of
Figure FDA0002429454960000034
Wherein the content of the first and second substances,
Figure FDA0002429454960000035
the euclidean distance of the kth load sample to the jth cluster center,
Figure FDA0002429454960000036
Figure FDA0002429454960000037
in step S33, each cluster center is iterated according to formula (4),
Figure FDA0002429454960000038
in step S34, the objective function f (h) is:
Figure FDA0002429454960000039
8. the method for studying spatial load distribution rules based on feature extraction according to claim 1, 6 or 7, wherein in the step S4,
and extracting load density distribution characteristics by using a nonparametric kernel density estimation method.
9. The method for studying spatial load distribution rules based on feature extraction as claimed in claim 8, wherein said step S4 specifically comprises:
s41, estimating nonparametric nuclear density;
and S42, extracting load density distribution characteristics according to the non-parameter kernel density estimation.
10. The method of claim 9, wherein the spatial load distribution rule is determined by a feature extraction method,
the kernel of the non-parametric kernel density estimation probability density function f (m) in step S41 is estimated as
Figure FDA0002429454960000041
Wherein n is the number of load samples, i.e. the number of electrically functional cells, m1,m2,L,mnFor the load density value, m, of each load sampleiThe load density value of the ith load sample.
CN202010231648.5A 2020-03-27 2020-03-27 Spatial load distribution rule research method based on feature extraction Pending CN111461197A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010231648.5A CN111461197A (en) 2020-03-27 2020-03-27 Spatial load distribution rule research method based on feature extraction

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010231648.5A CN111461197A (en) 2020-03-27 2020-03-27 Spatial load distribution rule research method based on feature extraction

Publications (1)

Publication Number Publication Date
CN111461197A true CN111461197A (en) 2020-07-28

Family

ID=71685709

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010231648.5A Pending CN111461197A (en) 2020-03-27 2020-03-27 Spatial load distribution rule research method based on feature extraction

Country Status (1)

Country Link
CN (1) CN111461197A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112232381A (en) * 2020-09-25 2021-01-15 国网上海市电力公司 Model parameter post-processing method and device for leading load parameter noise identification
CN112634078A (en) * 2020-12-18 2021-04-09 南京工程学院 Large-industrial load interruption priority evaluation method based on multi-dimensional index fusion
CN113077148A (en) * 2021-04-01 2021-07-06 四川中电启明星信息技术有限公司 Residential area openable capacity assessment method considering communication people flow data

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2009089594A (en) * 2007-09-28 2009-04-23 Kankoku Denryoku Kosha Temporal-spatial load analysis system of power facility utilizing inspection data and calculation method of load
CN103325071A (en) * 2013-06-09 2013-09-25 浙江大学 User typical load curve construction method based on density clustering
CN105426988A (en) * 2015-11-05 2016-03-23 国网福建省电力有限公司 Spacial load prediction method based on fuzzy rule
WO2016150407A1 (en) * 2015-03-23 2016-09-29 华南理工大学 Address resolution data-based construction land type rapid identification method
CN106022509A (en) * 2016-05-07 2016-10-12 国网浙江省电力公司经济技术研究院 Power distribution network space load prediction method taking region and load property dual differences into consideration
CN107169878A (en) * 2017-04-12 2017-09-15 华中科技大学 A kind of method based on information independence collection space load basic data of increasing income
CN108304978A (en) * 2018-05-08 2018-07-20 国网江西省电力有限公司经济技术研究院 A kind of mid-term Electric Power Load Forecast method based on data clusters theory
CN110717619A (en) * 2019-09-11 2020-01-21 国网浙江省电力有限公司经济技术研究院 Multi-scale space-time load prediction method and system for bottom-up power distribution network
CN110874381A (en) * 2019-10-30 2020-03-10 西安交通大学 User side load data abnormal value identification method based on space density clustering
US20200090285A1 (en) * 2018-03-16 2020-03-19 Dalian University Of Technology Method for short-term generation scheduling of cascade hydropower plants coupling cluster analysis and decision tree

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2009089594A (en) * 2007-09-28 2009-04-23 Kankoku Denryoku Kosha Temporal-spatial load analysis system of power facility utilizing inspection data and calculation method of load
CN103325071A (en) * 2013-06-09 2013-09-25 浙江大学 User typical load curve construction method based on density clustering
WO2016150407A1 (en) * 2015-03-23 2016-09-29 华南理工大学 Address resolution data-based construction land type rapid identification method
CN105426988A (en) * 2015-11-05 2016-03-23 国网福建省电力有限公司 Spacial load prediction method based on fuzzy rule
CN106022509A (en) * 2016-05-07 2016-10-12 国网浙江省电力公司经济技术研究院 Power distribution network space load prediction method taking region and load property dual differences into consideration
CN107169878A (en) * 2017-04-12 2017-09-15 华中科技大学 A kind of method based on information independence collection space load basic data of increasing income
US20200090285A1 (en) * 2018-03-16 2020-03-19 Dalian University Of Technology Method for short-term generation scheduling of cascade hydropower plants coupling cluster analysis and decision tree
CN108304978A (en) * 2018-05-08 2018-07-20 国网江西省电力有限公司经济技术研究院 A kind of mid-term Electric Power Load Forecast method based on data clusters theory
CN110717619A (en) * 2019-09-11 2020-01-21 国网浙江省电力有限公司经济技术研究院 Multi-scale space-time load prediction method and system for bottom-up power distribution network
CN110874381A (en) * 2019-10-30 2020-03-10 西安交通大学 User side load data abnormal value identification method based on space density clustering

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
刘思 等: "应用聚类分析与非参数核密度估计的空间负荷分布规律" *
肖白;刘庆永;房龙江;姚狄;宋凯豪;: "基于模糊粗糙集理论和时空信息的空间负荷预测" *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112232381A (en) * 2020-09-25 2021-01-15 国网上海市电力公司 Model parameter post-processing method and device for leading load parameter noise identification
CN112232381B (en) * 2020-09-25 2024-03-01 国网上海市电力公司 Model parameter post-processing method and device for dominant load parameter noise identification
CN112634078A (en) * 2020-12-18 2021-04-09 南京工程学院 Large-industrial load interruption priority evaluation method based on multi-dimensional index fusion
CN113077148A (en) * 2021-04-01 2021-07-06 四川中电启明星信息技术有限公司 Residential area openable capacity assessment method considering communication people flow data
CN113077148B (en) * 2021-04-01 2023-06-02 四川中电启明星信息技术有限公司 Residential area openable capacity assessment method considering communication people stream data

Similar Documents

Publication Publication Date Title
Thurstain‐Goodwin et al. Defining and delineating the central areas of towns for statistical monitoring using continuous surface representations
CN111461197A (en) Spatial load distribution rule research method based on feature extraction
CN111813835B (en) Public activity center recognition system based on mobile phone signaling and POI data
Wu et al. Modified data-driven framework for housing market segmentation
CN109359162B (en) GIS-based school site selection method
CN101315663A (en) Nature scene image classification method based on area dormant semantic characteristic
CN110119556B (en) Space-time evolution analysis method for regional water source conservation function
CN108256724B (en) Power distribution network open capacity planning method based on dynamic industry coefficient
CN106600063B (en) Space load prediction method based on fuzzy rough set theory and spatio-temporal information
CN112487724B (en) Urban dynamic expansion simulation method based on partition and improved CNN-CA model
CN108898244B (en) Digital signage position recommendation method coupled with multi-source elements
Mei et al. Population spatialization with pixel-level attribute grading by considering scale mismatch issue in regression modeling
Jia et al. Detecting the regional delineation from a network of social media user interactions with spatial constraint: A case study of Shenzhen, China
Sugianto et al. K-Means Algorithm For Clustering Poverty Data in Bangka Belitung Island Province
CN109684604A (en) A kind of city dynamic analysing method of the non-negative tensor resolution based on context-aware
Lin et al. An MIU-based deep embedded clustering model for urban functional zoning from remote sensing images and VGI data
Krumm et al. Land use inference from mobility traces
Li et al. Identifying urban form typologies in seoul with mixture model based clustering
CN114969143A (en) Inter-city atmospheric pollutant space-time correlation characteristic analysis method based on data mining
CN105808715B (en) Method for establishing map per location
CN115130579A (en) Population spatialization method and system for optimizing random forest based on improved clustering algorithm
CN111710157B (en) Method for extracting hot spot area of taxi
Zheng et al. Discovering urban functional regions with call detail records and points of interest: A case study of Guangzhou city
CN113393079A (en) Traffic zone dividing method based on public traffic data
Song et al. Unraveling the multi-scalar residential segregation and socio-spatial differentiation in China: A comparative study based on Nanjing and Hangzhou

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination